Merge pull request #1370 from cal-itp/ah_gtfs

Adding back rail routes
cal-itp · Feb 18, 2025 · 34830ba · 34830ba
2 parents 27991ba + cb5314b
commit 34830ba
Show file tree

Hide file tree

Showing 11 changed files with 30,391 additions and 192 deletions.
diff --git a/gtfs_digest/03_report.ipynb b/gtfs_digest/03_report.ipynb
diff --git a/gtfs_digest/45_missing_routes2.ipynb b/gtfs_digest/45_missing_routes2.ipynb
diff --git a/gtfs_digest/46_rail_routes.ipynb b/gtfs_digest/46_rail_routes.ipynb
diff --git a/gtfs_digest/47_check_operators.ipynb b/gtfs_digest/47_check_operators.ipynb
diff --git a/gtfs_funnel/route_typologies.py b/gtfs_funnel/route_typologies.py
@@ -222,10 +222,15 @@ def overlay_shapes_to_roads(
     buffer_meters: int
 ) -> gpd.GeoDataFrame:
 
+    # AH: removed pipe b/c it erases routes from Amtrak
+    #common_shape = gtfs_schedule_wrangling.most_common_shape_by_route_direction(
+    #    analysis_date
+    #).pipe(helpers.remove_shapes_outside_ca)
+
     common_shape = gtfs_schedule_wrangling.most_common_shape_by_route_direction(
         analysis_date
-    ).pipe(helpers.remove_shapes_outside_ca)
-
+    )
+    
     common_shape = common_shape.assign(
         route_meters = common_shape.geometry.length,
     )
@@ -352,6 +357,34 @@ def reconcile_route_and_nacto_typologies(
 
     return df2
 
+def add_rail_back(
+    categorize_routes_df: pd.DataFrame, overlay_shapes_to_roads_df: pd.DataFrame
+) -> pd.DataFrame:
+    """
+    categorize_routes_df: df created by categorize_routes_by_name()
+    overlay_shapes_to_roads_df: df created by overlay_shapes_to_roads() 
+    """
+    # Filter out for only rail routes and drop duplicates.
+    rail_routes = categorize_routes_df.loc[categorize_routes_df.is_rail == 1][
+        ["route_id", "schedule_gtfs_dataset_key"]
+    ].drop_duplicates()
+
+    # Merge with route_typologies_df to retain the details for
+    # columns such as typology, freq_category, etc
+    m1 = pd.merge(gdf, rail_routes, how="inner")
+
+    # Retain only one row for each route-direction-operator
+    # keeping the row with the highest pct_typology
+    m1 = m1.sort_values(
+        by=["route_id", "direction_id", "schedule_gtfs_dataset_key", "pct_typology"],
+        ascending=[True, True, True, False],
+    ).drop_duplicates(subset=["route_id", "direction_id", "schedule_gtfs_dataset_key"])
+
+    # Apply primary_secondary_typology() function which adds
+    # columns like is_nacto_rapid, is_nacto_coverage
+    m1 = primary_secondary_typology(m1)
+
+    return m1
 
 if __name__ == "__main__":
 
@@ -363,7 +396,7 @@ def reconcile_route_and_nacto_typologies(
 
     roads = delayed(prep_roads)(GTFS_DATA_DICT)
     ROAD_BUFFER_METERS = 20
-    TYPOLOGY_THRESHOLD = 0.10
+    TYPOLOGY_THRESHOLD = 0.1
 
     for analysis_date in analysis_date_list:
 
@@ -383,17 +416,33 @@ def reconcile_route_and_nacto_typologies(
         # Aggregate to route-dir-typology
         route_typology_df2 = primary_secondary_typology(route_typology_df)
 
+        # Tag if the route is express, rapid, or rail
         route_tagged = categorize_routes_by_name(analysis_date)
 
+        # Incorporate back rail routes that disappear if the routs
+        # dont't meet the minimum set in typology_threshold.
+        rail_routes_df = add_rail_back(route_tagged, gdf)
+        all_routes = pd.concat([route_typology_df2, rail_routes_df])
+
+
+        # Merge 
         df3 = pd.merge(
             route_tagged,
-            route_typology_df2,
+            all_routes,
             on = ["schedule_gtfs_dataset_key", "route_id"],
         ).pipe(reconcile_route_and_nacto_typologies)
 
-        df3.to_parquet(
-            f"{SCHED_GCS}{EXPORT}_{analysis_date}.parquet")
 
+        # Drop duplicates because some rail routes are found both
+        # route_typology_df2 and rail_routes_df
+        df3 = (df3.drop_duplicates(
+            subset = ["schedule_gtfs_dataset_key",
+                      "route_id", 
+                      "route_long_name", 
+                      "direction_id"])
+                      )
+        df3.to_parquet(
+            f"{SCHED_GCS}{EXPORT}_AH_TEST_{analysis_date}.parquet")
 
         time1 = datetime.datetime.now()
         print(f"route typologies {analysis_date}: {time1 - time0}")

diff --git a/gtfs_funnel/update_vars.py b/gtfs_funnel/update_vars.py
@@ -12,8 +12,8 @@
 )
 
 
-# analysis_date_list = [rt_dates.DATES["dec2024"]] + [rt_dates.DATES['nov2024']]
-analysis_date_list = all_dates
+analysis_date_list = [rt_dates.DATES["dec2024"]] + [rt_dates.DATES['nov2024']]
+# analysis_date_list = all_dates
 GTFS_DATA_DICT = catalog_utils.get_catalog("gtfs_analytics_data")
 
 COMPILED_CACHED_VIEWS = GTFS_DATA_DICT.gcs_paths.COMPILED_CACHED_VIEWS

diff --git a/portfolio/sites/gtfs_digest.yml b/portfolio/sites/gtfs_digest.yml
@@ -15,6 +15,7 @@ parts:
     - organization_name: Mendocino Transit Authority
     - organization_name: POINT
     - organization_name: Redwood Coast Transit Authority
+    - organization_name: Yurok Tribe
   - caption: District 02 - Redding
     params:
       district: 02 - Redding
@@ -94,6 +95,8 @@ parts:
       district: 05 - San Luis Obispo
     sections:
     - organization_name: City of Guadalupe
+    - organization_name: City of Lompoc
+    - organization_name: City of Morro Bay
     - organization_name: City of San Luis Obispo
     - organization_name: City of Santa Cruz
     - organization_name: City of Santa Maria
@@ -154,6 +157,7 @@ parts:
     - organization_name: City of Los Angeles
     - organization_name: City of Lynwood
     - organization_name: City of Maywood
+    - organization_name: City of Montebello
     - organization_name: City of Monterey Park
     - organization_name: City of Moorpark
     - organization_name: City of Norwalk
@@ -186,6 +190,7 @@ parts:
       district: 08 - San Bernardino
     sections:
     - organization_name: Basin Transit
+    - organization_name: Chemehuevi Indian Tribe
     - organization_name: City of Banning
     - organization_name: City of Beaumont
     - organization_name: City of Corona
@@ -194,6 +199,7 @@ parts:
     - organization_name: OmniTrans
     - organization_name: Palo Verde Valley Transit Agency
     - organization_name: Riverside Transit Agency
+    - organization_name: SunLine Transit Agency
     - organization_name: Victor Valley Transit Authority
   - caption: District 09 - Bishop
     params:
@@ -234,7 +240,13 @@ parts:
     params:
       district: 12 - Irvine
     sections:
+    - organization_name: Anaheim Transportation Network
+    - organization_name: City of Dana Point
+    - organization_name: City of Irvine
     - organization_name: City of Laguna Beach
+    - organization_name: City of Mission Viejo
+    - organization_name: City of San Clemente
+    - organization_name: City of San Juan Capistrano
     - organization_name: Orange County Transportation Authority
     - organization_name: University of California, Irvine
 readme: ./gtfs_digest/README.md

diff --git a/rt_scheduled_v_ran/logs/rt_v_scheduled_route_metrics.log b/rt_scheduled_v_ran/logs/rt_v_scheduled_route_metrics.log
@@ -103,3 +103,4 @@
 2025-01-16 15:49:56.361 | INFO     | __main__:route_metrics:88 - route aggregation 2023-10-11: 0:00:01.825395
 2025-01-16 15:49:58.178 | INFO     | __main__:route_metrics:88 - route aggregation 2023-11-15: 0:00:01.812722
 2025-01-16 15:50:00.055 | INFO     | __main__:route_metrics:88 - route aggregation 2023-12-13: 0:00:01.873527
+2025-01-21 16:29:01.096 | INFO     | __main__:route_metrics:88 - route aggregation 2024-11-13: 0:00:03.148850
diff --git a/rt_scheduled_v_ran/scripts/update_vars.py b/rt_scheduled_v_ran/scripts/update_vars.py
@@ -7,8 +7,8 @@
 oct2024_week = rt_dates.get_week("oct2024", exclude_wed=True)
 
 
-# analysis_date_list = [rt_dates.DATES["nov2024"]]
-analysis_date_list = rt_dates.y2024_dates + rt_dates.y2023_dates
+analysis_date_list = [rt_dates.DATES["nov2024"]]
+# analysis_date_list = rt_dates.y2024_dates + rt_dates.y2023_dates
 
 GTFS_DATA_DICT = catalog_utils.get_catalog("gtfs_analytics_data")
 

diff --git a/rt_segment_speeds/logs/avg_speeds.log b/rt_segment_speeds/logs/avg_speeds.log
@@ -678,3 +678,7 @@
 2025-01-16 16:28:31.739 | INFO     | __main__:summary_average_speeds:154 - rt_stop_times summary speed averaging for ['2023-04-15'] execution time: 0:00:17.603240
 2025-01-16 16:28:40.309 | INFO     | __main__:summary_average_speeds:120 - trip avg 0:00:08.508140
 2025-01-16 16:28:49.151 | INFO     | __main__:summary_average_speeds:154 - rt_stop_times summary speed averaging for ['2023-04-16'] execution time: 0:00:17.350031
+2025-01-21 16:31:27.163 | INFO     | __main__:summary_average_speeds:120 - trip avg 0:00:17.106380
+2025-01-21 16:31:41.624 | INFO     | __main__:summary_average_speeds:154 - rt_stop_times summary speed averaging for ['2024-12-11'] execution time: 0:00:31.567351
+2025-01-21 16:32:04.052 | INFO     | __main__:summary_average_speeds:120 - trip avg 0:00:22.327037
+2025-01-21 16:32:17.670 | INFO     | __main__:summary_average_speeds:154 - rt_stop_times summary speed averaging for ['2024-11-13'] execution time: 0:00:35.944913
diff --git a/rt_segment_speeds/segment_speed_utils/project_vars.py b/rt_segment_speeds/segment_speed_utils/project_vars.py
@@ -29,8 +29,9 @@
     rt_dates.oct2023_week, rt_dates.apr2023_week, 
 ]
 
-analysis_date_list = apr2024_week + oct2023_week + apr2023_week
-
+
+analysis_date_list = apr2024_week + oct2023_week + apr2023_week                     
+analysis_date_list = [rt_dates.DATES["dec2024"]] + [rt_dates.DATES['nov2024']]
 
 PROJECT_CRS = "EPSG:3310"
 ROAD_SEGMENT_METERS = 1_000